Working with Rayven’s MySQL Tables in The Workflow

Rayven’s platform provides powerful, native support for MySQL-based tables that serve as the foundation for storing, querying, and managing structured data in workflows and dashboards.

Whether you're building real-time logic or configuring dashboards, Rayven tables provide the reliable backend needed for persistent data operations.

Table Types and Structure

Rayven supports two types of tables: Primary and Secondary.

1. Primary Tables

• Used for persistent, long-term storage.

• Suitable for raw data, logs, calculated metrics, and historical events.

• Fully read/write capable, with support for inserts, updates, and upserts.

2. Secondary Tables

• Used for reference data, configuration values, and cached responses.

• Typically read-heavy, updated less frequently.

• Ideal for storing lookups (e.g., unit conversions, device mappings).

💡 Tip: Choose based on usage—primary for high-volume transactional data; secondary for lightweight configs or lookups.

Creating Tables

Tables can be created via the Data Tables interface.

Manual Table Creation

1. Navigate to the Tables section in the Data menu.

2. Click “+ New Table”.

3. Define the table name, schema (fields, types), and description.

4. Save and publish the structure.

Auto-Creation via Templates

• Rayven provides prebuilt templates to auto-generate commonly used tables (e.g., device logs, alert tracking).

• Once created, tables can be updated, but cannot be deleted, so design carefully.

Writing to Tables in Workflows

To store data in a table:

1. 1. Drag in the Update Table Node.

   2. Select the target table.

   3. Choose the key field (e.g., device_id, uid) for row matching.

   4. Map workflow variables to table columns.

   5. Set behavior:

      • Insert only: New records only

      • Update if exists: Modify existing rows

      • Upsert: Insert or update depending on presence

Reading from Tables in Workflows

To retrieve data:

1. Use the Query Table Node.

2. Select the table.

3. Apply filter conditions using:

   • Static values (e.g., status = “active”)

   • Dynamic variables from workflow context

4. Set result mode:

   • First match

   • Multiple rows

   • All rows (use cautiously)

Output data can be:

• Used in logic nodes

• Injected into HTML Nodes for dynamic interface display

Table Labels and Filtering

What Are Labels?

Labels are user-defined tags attached to table rows to organize and filter data.

Examples:

• region = north

• priority = high

• device_type = pump

Applying Labels

• Manually in the table UI

• Automatically within a workflow using the Table Update node

Using Labels

Query Patterns & Optimization

• Use indexed fields (IDs, timestamps) for performance.

• Avoid full-table reads unless necessary—always filter where possible.

• Use secondary tables for reference data to avoid repeated API calls.

• Combine label filters and UID variables for granular results.

Best Practices

• Lock schema in production tables to prevent accidental changes.

• Use consistent field names (e.g., device_id, event_date).

• Regularly archive or purge large primary tables to maintain performance.

• Document table purposes, schema definitions, and label conventions.

• Use role-based access control to protect sensitive data.

Q&A

Q: Can I update multiple rows in a table at once?
A: No. Table updates are performed on a per-row basis using a defined key field. To affect multiple rows, loop logic is needed.

Q: What happens if the upsert key field is missing?
A: The row will not update or insert properly. Always ensure that key fields like device_id or uid are populated before the update.

Q: Can I read from multiple tables in the same workflow?
A: Yes. Use separate Query Table Nodes for each table, and merge or compare results using logic or expression nodes.

Q: Can I create relationships between tables?
A: Not directly like SQL JOINs, but you can simulate lookups by querying one table and using the result to filter a second query.

Q: Can I use table data in HTML or dashboards?
A: Absolutely. Query results can be injected into HTML Nodes or used to populate form controls, widgets, or visualizations.

Q: Are table updates real-time?
A: Yes. Table writes occur instantly within the workflow execution context and are available for immediate downstream use.